Peroxide is the most commonly used catalyst in the polymerization of polystyrene which causes free radicals to be generated to initiate the polymerization reaction. The polystyrene produced from the peroxide-catalyzed processes belongs to the type of atactic polystyrene (aps), which, by definition, does not possess any stereo regularity. The atactic polystyrene, which has been widely used in many commercial applications for more half century, is an amorphous polymer. The amorphous physical characteristics of atactic polystyrene have limited the range of their applications. The atactic polystyrene is primarily used in a relatively low-value added market, and they typically cannot be used in engineering plastics applications.
Isotactic polystyrene (ips), on the other hand, was also developed in as early as 1955, by G. Natta using the co-called Ziegler-Natta catalyst. The isotactic polystyrene are a highly crystalline polymer, and it exhibits a very high melting point (240.degree. C.). These properties make the isotactic polystyrene a suitable candidate for many engineering plastics applications. However, the isotactic polystyrene suffers from the problem of having undesirably low crystallization rate, thus causing fabrication difficulties. Unless this problem can be overcome, the isotactic polystyrene does not appear to have very high commercial potential.
Relative to atactic polystyrene and isotactic polystyrene, the syndiotactic polystyrene was relatively late comer. It was not until 1986 when the syndiotactic polystyrene was first developed by Ishihara using a metallocene catalyst composition. Typically, the polymerization of syndiotactic polystyrene requires a catalyst composition containing a transitional metal titanium complex and methyl aluminoxane (or "MAO"). The concerted actions of the titanium complex and the methyl aluminoxane allows syndiotactic polystyrene to be polymerized. Descriptions of the processes for preparing syndiotactic polystyrene have been provided in, for example, European Patent Application EP 210,615, in which a catalyst composition containing tetra(ethyoxy)titanium and methyl aluminoxane was used for preparing syndiotactic polystyrene; and in world patent application WO 8,810,275, in which high syndiotacticity polystyrene was reported to have been prepared using a catalyst composition containing cyclopentadienyl trichlorotitanium and methyl aluminoxane.
U.S. Pat. Nos. 4,774,301 and 4,808,680 disclosed the use of a catalyst composition containing a transitional metal zirconium complex and methyl aluminoxane for preparing syndiotactic polystyrene. Compared to the catalyst compositions using a titanium complex and methyl aluminoxane, these catalyst compositions containing the zirconium complex exhibited noticeably lower activity, and the polystyrene so produced exhibited relatively lower molecular weight and lower degree of syndiotacticity.
All of the catalyst compositions described above for preparing syndiotactic polystyrene contain Group IV transitional metal complexes and methyl aluminoxane to provide activation. It should be noted that a very high excess of methyl aluminoxane is required to provide the desired activated catalyst. Because of the high cost of methyl aluminoxane, these processes have very limited commercial applications. Thus it is highly desirable to develop a metallocene based catalyst composition which can minimize, or even eliminate, the amount of methyl aluminoxane required. European Patent Application 505,890 and World Patent Application WO 930,367 disclosed a catalyst composition, which contains cyclopentadienyl trialkyl titanium as a catalyst, a non-coordinated Lewis acid as a co-catalyst, and triisobutyl aluminum as a scavenger, for the preparation of high syndiotactic polystyrene. These catalyst compositions, however, have relatively low activity.
Within the family of titanium complexes, or more specifically titanocenes, the catalytic activity, for polymerizing polystyrene, is higher for titanocenes containing one cyclopentadienyl ligand than those titanocenes containing two cyclopentadienyl ligands. The catalytic activity of the titanocene containing one cyclopentadienyl ligand is also higher than titanium complexes containing no cyclopentadienyl ligand (which is thus by definition not a titanocene). This relative relationship has been disclosed in European Patent Application EP 210,615. While the catalytic activity of the titanium complexes containing no cyclopentadienyl ligand is substantially lower than the titanocene containing one cyclopentadienyl ligand, it has the advantage of being substantially cheaper. Therefore, from economic considerations, it is highly desirable to develop a co-catalyst composition that can substantially increase the activity of the cheapest titanium complexes, that is, the titanium complexes that contain no cyclopentadienyl ligand, so as to lower the overall cost of the metallocene catalyst composition while providing excellent catalytic activity.